> long thought piece : bodies, data, motion capture, aiaiai
For *Where Things Go*, @ msomil and I have been experimenting with ways to capture movement and motion to visualize it in live performance. We work through the various attributes of choreographed movement, sometimes abstracting it to the physics of the motion and more so now working with full body three-dimensional representations.
The pipeline remains more or less the same: dancer with or without a motion capture suit, data capture, skeletal reconstruction, cleanup and animation. With advances in motion capture tools like Move AI, QuickMagic,..., the process becomes accessible and eerily simpler[?] in some ways. With the pace of these new tools, the possibilities open up, especially for the animation field.
@dancemagazine recently covered how AI motion capture is "making animating movement quicker, cheaper, and more accessible." If you squint though, here's what happens, especially in the most accessible free tier - your video gets processed and the movement is mapped onto a stock motion mesh. Probably modeled into a "generic" skeleton or a robot figure. You're reconfiguring your body to fit a standardized mesh. This is where the interrogation in process lies.
Ten years ago [whaa], I was building motion capture repositories to analyze and pull out unique patterns in individual movement (swipe to #4 ✨). The central tension: how do you retain or highlight specificity in a system designed to flatten difference? Rewind more to Ghostcatching (1999) by Bill T. Jones, Paul Kaiser and Shelley Eshkar. Ghostcatching captured Bill T. Jones's dancing body and transformed it into 3D drawings. They asked: "What is human movement abstracted from its original body? Can the drawn line carry the rhythm, weight, and intent of physical movement?"
✨ These interrogations, still stand > especially as the tools are continuously evolving. The accessibility matters, but so do the questions: What gets normalized in the name of that ease? Whose proportions are baked into the standardized digital avatars? Whose idea of movement or dance? How does technology impose its logic on how dance is seen, performed and remembered?